Distinguishing Natural and AI-generated Images (DNAI) dataset
收藏资源简介:
DNAI数据集是由中山大学和新加坡国立大学等机构创建的一个大规模多模态数据集,旨在评估AI生成图像与自然图像之间的差异。该数据集包含超过44万张由8种代表性生成模型生成的AI图像,使用文本到图像(T2I)、图像到图像(I2I)和文本与图像到图像(TI2I)等多种提示方式生成。数据集的创建过程结合了多种生成模型和多模态提示,确保了数据的多样性和广泛性。该数据集主要应用于AI生成图像的质量评估,旨在解决AI生成图像与自然图像之间的差异问题,推动AI生成图像在实际应用中的发展。
The DNAI Dataset is a large-scale multimodal dataset created by institutions including Sun Yat-sen University and the National University of Singapore, aiming to evaluate the discrepancies between AI-generated images and natural images. This dataset contains over 440,000 AI-generated images produced by 8 representative generative models through multiple prompting methods such as text-to-image (T2I), image-to-image (I2I), and text-and-image-to-image (TI2I). The dataset construction process integrates various generative models and multimodal prompts, ensuring the diversity and broad coverage of the collected data. It is primarily applied to the quality assessment of AI-generated images, with the objective of addressing the differences between AI-generated and natural images and advancing the practical development and real-world application of AI-generated images.
ANID 数据集概述
数据集名称
- ANID
数据集描述
- 暂无详细描述,内容为 "Come soon!"。

- 1ANID: How Far Are We? Evaluating the Discrepancies Between AI-synthesized Images and Natural Images through Multimodal Guidance中山大学网络空间安全学院,新加坡国立大学,云南大学网络空间研究中心 · 2024年



